Do seasonal profiles of foliar pigments improve species discrimination of evergreen coastal tree species in KwaZulu- Natal, South Africa?

Abstract

Studies in the Northern hemisphere have shown the potential of foliar pigment seasonal profiles as a means of improving species discrimination. Remote sensing vegetation indices have been used to optimise absorption features presented by foliar pigments, as well as improve species discrimination. This study investigated the potential of seasonal pigment profiles (for foliar carotenoid and total chlorophyll) in improving species discrimination for trees using leaf spectral data. Our aims were to (i) determine whether species have unique seasonal profiles of carotenoids and chlorophyll; and (ii) whether these seasonal profiles can be used to improve species discrimination, compared to single season pigment concentrations. We sampled sunlit leaves of seven evergreen tree species in a sub-tropical region of South Africa, over four seasons during 2011-12. Parametric ANOVA classification was compared to similarity measures of shape (spectral angle mapper; SAM) and magnitude (sum of Euclidean Distance; ED). For both pigments, the parametric analysis of combined seasonal content improved species discrimination when compared to single season content and the similarity measures. ED outperformed SAM in species discrimination for both pigments. Multi-seasonal carotenoid and chlorophyll content information improved species discrimination of evergreen coastal tree species in KwaZulu-Natal, South Africa.